dsl.py 15.6 KB
Newer Older
1
"""dsl.py - Support for domain specific notations for DHParser
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

Copyright 2016  by Eckhart Arnold (arnold@badw.de)
                Bavarian Academy of Sciences an Humanities (badw.de)

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at

    http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
implied.  See the License for the specific language governing
permissions and limitations under the License.

Module ``DSLsupport`` contains various functions to support the
compilation of domain specific languages based on an EBNF-grammar.
"""

di68kap's avatar
di68kap committed
22
import collections
23
import os
24

25 26 27 28 29
try:
    import regex as re
except ImportError:
    import re

30
from .ebnf import EBNFGrammar, EBNFTransform, EBNFCompiler, grammar_changed
31 32
from .toolkit import load_if_file, is_python_code, compile_python_object
from .parsers import GrammarBase, CompilerBase, full_compilation, nil_scanner
33
from .syntaxtree import Node
34

Eckhart Arnold's avatar
Eckhart Arnold committed
35 36 37 38 39

__all__ = ['GrammarError',
           'CompilationError',
           'load_compiler_suite',
           'compileDSL',
40
           'compile_on_disk']
Eckhart Arnold's avatar
Eckhart Arnold committed
41 42


43 44 45 46 47 48 49 50 51
SECTION_MARKER = """\n
#######################################################################
#
# {marker}
#
#######################################################################
\n"""

RX_SECTION_MARKER = re.compile(SECTION_MARKER.format(marker=r'.*?SECTION.*?'))
52
RX_WHITESPACE = re.compile('\s*')
53 54 55 56 57 58

SYMBOLS_SECTION = "SYMBOLS SECTION - Can be edited. Changes will be preserved."
SCANNER_SECTION = "SCANNER SECTION - Can be edited. Changes will be preserved."
PARSER_SECTION = "PARSER SECTION - Don't edit! CHANGES WILL BE OVERWRITTEN!"
AST_SECTION = "AST SECTION - Can be edited. Changes will be preserved."
COMPILER_SECTION = "COMPILER SECTION - Can be edited. Changes will be preserved."
di68kap's avatar
di68kap committed
59
END_SECTIONS_MARKER = "END OF DHPARSER-SECTIONS"
60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83


class GrammarError(Exception):
    """Raised when (already) the grammar of a domain specific language (DSL)
    contains errors.
    """

    def __init__(self, error_messages, grammar_src):
        self.error_messages = error_messages
        self.grammar_src = grammar_src


class CompilationError(Exception):
    """Raised when a string or file in a domain specific language (DSL)
    contains errors.
    """

    def __init__(self, error_messages, dsl_text, dsl_grammar, AST):
        self.error_messages = error_messages
        self.dsl_text = dsl_text
        self.dsl_grammar = dsl_grammar
        self.AST = AST

    def __str__(self):
84
        return '\n'.join(self.error_messages)
85 86


87
DHPARSER_IMPORTS = '''
di68kap's avatar
di68kap committed
88
from functools import partial
89
import sys
di68kap's avatar
di68kap committed
90 91 92 93
try:
    import regex as re
except ImportError:
    import re
94
from DHParser.toolkit import load_if_file    
95
from DHParser.parsers import GrammarBase, CompilerBase, nil_scanner, \\
di68kap's avatar
di68kap committed
96 97
    Lookbehind, Lookahead, Alternative, Pop, Required, Token, \\
    Optional, NegativeLookbehind, OneOrMore, RegExp, Retrieve, Sequence, RE, Capture, \\
98
    ZeroOrMore, Forward, NegativeLookahead, mixin_comment, full_compilation
99 100 101 102
from DHParser.syntaxtree import Node, traverse, remove_enclosing_delimiters, \\
    remove_children_if, reduce_single_child, replace_by_single_child, remove_whitespace, \\
    no_operation, remove_expendables, remove_tokens, flatten, is_whitespace, is_expendable, \\
    WHITESPACE_KEYWORD, TOKEN_KEYWORD
103
'''
di68kap's avatar
di68kap committed
104

105

106 107 108 109
DHPARSER_COMPILER = '''
def compile_{NAME}(source):
    """Compiles ``source`` and returns (result, errors, ast).
    """
110
    return full_compilation(source, {NAME}Scanner,
111
        {NAME}Grammar(), {NAME}Transform, {NAME}Compiler())
112 113 114 115 116 117 118 119 120 121 122 123 124 125

if __name__ == "__main__":
    if len(sys.argv) > 1:
        result, errors, ast = compile_{NAME}(sys.argv[1])
        if errors:
            for error in errors:
                print(error)
                sys.exit(1)
        else:
            print(result)
    else:
        print("Usage: {NAME}_compiler.py [FILENAME]")
'''

126

127
def grammar_instance(grammar_representation):
128 129 130 131 132
    """Returns a grammar object and the source code of the grammar, from
    the given `grammar`-data which can be either a file name, ebnf-code,
    python-code, a GrammarBase-derived grammar class or an instance of
    such a class (i.e. a grammar object already).
    """
133
    if isinstance(grammar_representation, str):
134
        # read grammar
135 136
        grammar_src = load_if_file(grammar_representation)
        if is_python_code(grammar_representation):
137 138
            parser_py, errors, AST = grammar_src, '', None
        else:
139
            parser_py, errors, AST = full_compilation(grammar_src, None,
140
                                                      EBNFGrammar(), EBNFTransform, EBNFCompiler())
141
        if errors:
142
            raise GrammarError('\n\n'.join(errors), grammar_src)
di68kap's avatar
di68kap committed
143
        parser_root = compile_python_object(DHPARSER_IMPORTS + parser_py, '\w*Grammar$')()
144 145 146
    else:
        # assume that dsl_grammar is a ParserHQ-object or Grammar class
        grammar_src = ''
147 148
        if isinstance(grammar_representation, GrammarBase):
            parser_root = grammar_representation
149 150
        else:
            # assume `grammar` is a grammar class and get the root object
151
            parser_root = grammar_representation()
152 153 154
    return parser_root, grammar_src


155
def compileDSL(text_or_file, scanner, dsl_grammar, ast_transformation, compiler):
156
    """Compiles a text in a domain specific language (DSL) with an
157 158
    EBNF-specified grammar. Returns the compiled text or raises a
    compilation error.
159 160 161
    
    Raises:
        CompilationError if any errors occured during compilation
162 163 164
    """
    assert isinstance(text_or_file, str)
    assert isinstance(compiler, CompilerBase)
165
    parser_root, grammar_src = grammar_instance(dsl_grammar)
166
    src = load_if_file(text_or_file)
167 168
    result, errors, AST = full_compilation(src, scanner, parser_root,
                                           ast_transformation, compiler)
169
    if errors:  raise CompilationError(errors, src, grammar_src, AST)
170 171 172
    return result


173
def compileEBNF(ebnf_src, ebnf_grammar_obj=None, source_only=False):
174 175 176 177 178 179 180
    """Compiles an EBNF source file into a Grammar class.

    Please note: This functions returns a class which must be 
    instantiated before calling its parse()-method! Calling the method
    directly from the class (which is technically possible in python
    yields an error message complaining about a missing parameter,
    the cause of which may not be obvious at first sight. 
181 182 183 184 185 186 187 188

    Args:
        ebnf_src(str):  Either the file name of an EBNF grammar or
            the EBNF grammar itself as a string.
        ebnf_grammar_obj:  An existing instance of the 
            DHParser.EBNFcompiler.EBNFGrammar object. This can speed
            up compilation, because no new EBNFGrammar object needs to
            be instantiated.
189 190
        source_only (bool):  If True, the source code of the Grammar
            class is returned instead of the class itself.
191 192
    Returns:
        A Grammar class that can be instantiated for parsing a text
193
        which conforms to the language defined by ``ebnf_src``.
194 195
    """
    grammar = ebnf_grammar_obj or EBNFGrammar()
196
    grammar_src = compileDSL(ebnf_src, nil_scanner, grammar, EBNFTransform, EBNFCompiler())
197 198
    return grammar_src if source_only else \
        compile_python_object(DHPARSER_IMPORTS + grammar_src, '\w*Grammar$')
199 200


201 202 203 204 205 206 207 208 209 210 211 212 213 214 215
def load_compiler_suite(compiler_suite):
    """Extracts a compiler suite from file or string ``compiler suite``
    and returns it as a tuple (scanner, parser, ast, compiler).
    """
    global RX_SECTION_MARKER
    assert isinstance(compiler_suite, str)
    source = load_if_file(compiler_suite)
    if is_python_code(compiler_suite):
        try:
            intro, imports, scanner_py, parser_py, ast_py, compiler_py, outro = \
                RX_SECTION_MARKER.split(source)
        except ValueError as error:
            raise AssertionError('File "' + compiler_suite + '" seems to be corrupted. '
                                 'Please delete or repair file manually.')
        scanner = compile_python_object(imports + scanner_py, '\w*Scanner$')
216
        ast = compile_python_object(imports + ast_py, '\w*Transform$')
217 218 219 220
        compiler = compile_python_object(imports + compiler_py, '\w*Compiler$')
    else:
        # assume source is an ebnf grammar
        parser_py, errors, AST = full_compilation(
221
            source, None, EBNFGrammar(), EBNFTransform, EBNFCompiler())
222 223 224
        if errors:
            raise GrammarError('\n\n'.join(errors), source)
        scanner = nil_scanner
225
        ast = EBNFTransform
226 227
        compiler = EBNFCompiler
    parser = compile_python_object(DHPARSER_IMPORTS + parser_py, '\w*Grammar$')
228 229 230 231 232 233 234 235 236 237 238 239 240

    return scanner, parser, ast, compiler


def suite_outdated(compiler_suite, grammar_source):
    """Returns ``True``  if the ``compile_suite`` needs to be updated.
     
    An update is needed, if either the grammar in the compieler suite 
    does not reflect the latest changes of ``grammar_source`` or if 
    sections from the compiler suite have diligently been overwritten
    with whitespace order to trigger their recreation. Note: Do not
    delete or overwrite the section marker itself. 

241
    Args:
242 243 244 245 246 247 248 249 250 251 252 253 254 255 256
        compiler_suite:  the parser class representing the grammar
            or the file name of a compiler suite containing the grammar
        grammar_source:  File name or string representation of the
            EBNF code of the grammar

    Returns (bool):
        True, if ``compiler_suite`` seems to be out of date.
    """
    try:
        scanner, grammar, ast, compiler = load_compiler_suite(compiler_suite)
        return grammar_changed(grammar, grammar_source)
    except ValueError:
        return True


257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280
def run_compiler(text_or_file, compiler_suite):
    """Compiles a source with a given compiler suite.

    Args:
        text_or_file (str):  Either the file name of the source code or
            the source code directly. (Which is determined by 
            heuristics. If ``text_or_file`` contains at least on
            linefeed then it is always assumed to be a source text and
            not a file name.)
        compiler_suite(str):  File name of the compiler suite to be
            used.
        
    Returns:
        The result of the compilation, the form and type of which 
        depends entirely on the compiler.
        
    Raises:
        CompilerError
    """
    scanner, parser, ast, compiler = load_compiler_suite(compiler_suite)
    return compileDSL(text_or_file, scanner, parser(), ast, compiler())


def compile_on_disk(source_file, compiler_suite="", extension=".xml"):
281 282 283
    """Compiles the a source file with a given compiler and writes the
    result to a file.

Eckhart Arnold's avatar
Eckhart Arnold committed
284 285 286 287 288 289
    If no ``compiler_suite`` is given it is assumed that the source
    file is an EBNF grammar. In this case the result will be a Python
    script containing a parser for that grammar as well as the
    skeletons for a scanner, AST transformation table, and compiler.
    If the Python script already exists only the parser name in the
    script will be updated. (For this to work, the different names
290
    need to be delimited section marker blocks.). `compile_on_disk()`
Eckhart Arnold's avatar
Eckhart Arnold committed
291 292
    returns a list of error messages or an empty list if no errors
    occurred.
293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309
    
    Parameters:
        source_file(str):  The file name of the source text to be
            compiled.
        compiler_suite(str):  The file name of the compiler suite
            (usually ending with '_compiler.py'), with which the source
            file shall be compiled. If this is left empty, the source
            file is assumed to be an EBNF-Grammar that will be compiled
            with the internal EBNF-Compiler.
        extension(str):  The result of the compilation (if successful)
            is written to a file with the same name but a different
            extension than the source file. This parameter sets the
            extension.
            
    Returns:
        A list of error messages or an empty list if there were no 
        errors. 
Eckhart Arnold's avatar
Eckhart Arnold committed
310
    """
311
    filepath = os.path.normpath(source_file)
312 313
    # with open(source_file, encoding="utf-8") as f:
    #     source = f.read()
314
    rootname = os.path.splitext(filepath)[0]
315
    compiler_name = os.path.basename(rootname)
316
    if compiler_suite:
317 318
        scanner, pclass, trans, cclass = load_compiler_suite(compiler_suite)
        parser = pclass()
319
        compiler1 = cclass()
320 321 322
    else:
        scanner = nil_scanner
        parser = EBNFGrammar()
323
        trans = EBNFTransform
324 325
        compiler1 = EBNFCompiler(compiler_name, source_file)
    result, errors, ast = full_compilation(source_file, scanner, parser, trans, compiler1)
326 327 328
    if errors:
        return errors

329
    elif trans == EBNFTransform:  # either an EBNF- or no compiler suite given
330 331
        global SECTION_MARKER, RX_SECTION_MARKER, SCANNER_SECTION, PARSER_SECTION, \
            AST_SECTION, COMPILER_SECTION, END_SECTIONS_MARKER
332
        f = None
333 334 335
        try:
            f = open(rootname + '_compiler.py', 'r', encoding="utf-8")
            source = f.read()
336 337
            sections = RX_SECTION_MARKER.split(source)
            intro, imports, scanner, parser, ast, compiler, outro = sections
338
        except (PermissionError, FileNotFoundError, IOError) as error:
339
            intro, imports, scanner, parser, ast, compiler, outro = '', '', '', '', '', '', ''
340 341 342 343
        except ValueError as error:
            raise ValueError('File "' + rootname + '_compiler.py" seems to be corrupted. '
                                                   'Please delete or repair file manually!')
        finally:
344 345 346
            if f:
                f.close()
                f = None
347

348 349 350 351 352 353 354 355 356 357 358 359 360
        if RX_WHITESPACE.fullmatch(intro):
            intro = '#!/usr/bin/python'
        if RX_WHITESPACE.fullmatch(outro):
            outro = DHPARSER_COMPILER.format(NAME=compiler_name)
        if RX_WHITESPACE.fullmatch(imports):
            imports = DHPARSER_IMPORTS
        if RX_WHITESPACE.fullmatch(scanner):
            scanner = compiler1.gen_scanner_skeleton()
        if RX_WHITESPACE.fullmatch(ast):
            ast = compiler1.gen_AST_skeleton()
        if RX_WHITESPACE.fullmatch(compiler):
            compiler = compiler1.gen_compiler_skeleton()

361 362 363 364
        try:
            f = open(rootname + '_compiler.py', 'w', encoding="utf-8")
            f.write(intro)
            f.write(SECTION_MARKER.format(marker=SYMBOLS_SECTION))
365
            f.write(imports)
366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398
            f.write(SECTION_MARKER.format(marker=SCANNER_SECTION))
            f.write(scanner)
            f.write(SECTION_MARKER.format(marker=PARSER_SECTION))
            f.write(result)
            f.write(SECTION_MARKER.format(marker=AST_SECTION))
            f.write(ast)
            f.write(SECTION_MARKER.format(marker=COMPILER_SECTION))
            f.write(compiler)
            f.write(SECTION_MARKER.format(marker=END_SECTIONS_MARKER))
            f.write(outro)
        except (PermissionError, FileNotFoundError, IOError) as error:
            print('# Could not write file "' + rootname + '_compiler.py" because of: '
                  + "\n# ".join(str(error).split('\n)')))
            print(result)
        finally:
            if f:  f.close()

    else:
        try:
            f = open(rootname + extension, 'w', encoding="utf-8")
            if isinstance(result, Node):
                f.write(result.as_xml())
            else:
                f.write(result)
        except (PermissionError, FileNotFoundError, IOError) as error:
            print('# Could not write file "' + rootname + '.py" because of: '
                  + "\n# ".join(str(error).split('\n)')))
            print(result)
        finally:
            if f:  f.close()

    return []